276 



Prof. Karl Pearson. 



munity under one system, but upon all the data which, for some time 

 to come, appear likely to be available. These data are neither quan- 

 titatively nor qualitatively ideal, but, on the other hand, they must 

 be given a reasonable amount of weight in considering whether, at 

 any rate in the case of one organ stature, any steady telegenic 

 influence can be traced in man. 



The reduction from the family measurement-cards, the formation 

 of the eight correlation tables, and the calculation of both variation 

 and correlation coefficients have been undertaken by Miss Alice Lee 

 of Bedford College, a task requiring much labour and persistency. 

 I have independently verified, and in some minor points corrected 

 her calculations, as well as added the probable errors of the con- 

 stants determined. 



(3) The following are the means and standard-deviations with 

 their probable errors for the various groups. 



Table I. Stature of Families in Inches. 



All the quantities have here been calculated precisely as in my 

 third memoir on the mathematical theory of evolution (see ' Phil. 

 Trans./ A, vol. 187, pp. 270 271). In this case, however, no child is 

 included twice as a child, and parents are not weighted with their 

 offspring. Thus reproductive selection is not allowed to influence the 

 results. 



It will be seen that the probable errors of the means and standard 

 deviations are, as in the former paper, too large to allow of 

 absolutely definite conclusions when those conclusions are not sup- 

 ported by a continuous change of values, or directly verified by the 

 numbers of the earlier memoir. But one or two such conclusions 

 may be drawn, and I will note them before passing to correlation. 



(i) The law of sexual interchange referred to in my former paper 

 (p. 274) is confirmed with greater uniformity. Fathers of sons are 

 sensibly less variable than fathers of daughters, and mothers of 

 daughters are sensibly less variable than mothers of sons. In other 



